Strain monitoring method based on conditional generative adversarial network and load strain linear superposition

Abstract

The invention provides a strain monitoring method based on a conditional generative adversarial network and load strain linear superposition. The method comprises the steps of carrying out the static simulation of a structure, and obtaining the simulation strain data of the structure; according to the simulated sensor layout, obtaining strain real data of each sensor on the structure, obtaining a strain column vector of a measuring point on the structure, further obtaining a strain weight, generating an adversarial network by using a condition, enabling the simulated strain data to learn the real strain data, and obtaining pseudo experiment strain data under a working condition; obtaining a relation between a strain column vector of a measuring point and a model error through pseudo experiment strain data by utilizing an extreme learning machine, and carrying out model error correction; and according to the strain matrixes and the strain weights of all the points of the structure, obtaining strain values of all the points on the structure in combination with an error correction result. According to the invention, the difference between the simulation model and the experimental model is reduced by using the conditional generative adversarial network, and the strain field reconstruction of the structure is realized through a load strain linear superposition algorithm.

Description

technical field [0001] The invention relates to the technical field of strain monitoring, in particular to a strain monitoring method in which a condition-generating confrontation network and a load-strain linear superimposition are performed. Background technique [0002] The statements in this section merely provide background art related to the present invention and do not necessarily constitute prior art. [0003] The underframe crossbeam is one of the key structural parts of the train. It is connected by bolts to hang brake equipment, air-conditioning equipment and other under-car equipment weighing several tons. When the train is running at high speed, it bears the vertical static load brought by the equipment under the train. At the same time, due to the interaction between the track and the train, the equipment under the train will vibrate in the vertical direction, so that the beam bears the dynamic load. The combined effect of various complex loads can easily caus...

AI Technical Summary

Problems solved by technology

The modal method is the most widely used, although the method is fast and simple to calculate, but because the modal method requires the strain mode shape and displacement mode shape of the structure, the selection of the mode shape order is particularly important, and not all modes in engineering Can be excited, the mode needs to be intercepted during the calculation, and the order of the intercepted mode directly affects the final calculation result, the mode order is difficult to determine, and the complex structure mode test is difficult, and the reconstruction effect is difficult ensure

Images